#29 - 2019.11 - Spacy and Transformers; Applied Neural Machine Translation

Hacking Machine Learning
Hacking Machine Learning
Public group
Location image of event venue


🆂🅻🅾🆃 #1: spaCy meets Transformers - Ines Montani, Founder of https://spacy.io and https://prodi.gy

Huge transformer models like BERT, GPT-2 and XLNet have set a new standard for accuracy on almost every Natural Language Processing leaderboard. However, these models are very new, and most of the software ecosystem surrounding them is oriented towards the many opportunities for further research that they provide. In this talk, I'll describe how you can now use these models in spaCy, an open-source library for putting Natural Language Processing to work on real problems. I'll also discuss the many opportunities that new transfer learning technologies can offer production NLP, regardless of which specific software packages you choose to get the job done.

Ines is a developer specializing in tools for AI and NLP technology. She’s the co-founder of Explosion and a core developer of spaCy, a popular open-source library for Natural Language Processing in Python, and Prodigy, a modern annotation tool for creating training data for machine learning models.

🆂🅻🅾🆃 #2: Adding Neural Machine Translation to a Word Processor - Thomas Viehmann

Promising to fulfill the human dream of understanding something written in a foreign language without learning it,
Neural Machine Translation is one of the poster applications in Natural Language Processing. We all know the impressive
results of online services. Here, we look at how to start from OpenNMT - a popular open-source NMT framework - and take
it all the way to production as an extension for the LibreOffice Writer that runs fully on your PC.

We will take a look at the OpenNMT library, the translation models it implements, how to train them, and what it takes
to use it from LibreOffice, resulting in LibreOffice Translate available at https://github.com/lernapparat/lotranslate. We also highlight some recent developments around distillation and how we can use them to improve LibreOffice Translate.

Thomas is a machine learning specialty trainer and consultant with a focus on PyTorch at MathInf GmbH. He is a PyTorch core developer and
has contributed over 100 features and bug fixes across many parts of PyTorch. He blogs about Machine Learning at https://lernapparat.de/

📌As always, we are super happy to have you with a lightning talk on your small or big successes, open source projects or just something amazing you would like to share.

📌Do not miss out updates on twitter - https://twitter.com/hack_ai and join Munich ML community on Slack channel https://ai-hack-inviter.herokuapp.com/